java add character into a string

The problem is not the isDigit detection, the problem is builder.insert(i, "\n");!

try this, it works without insert:

@org.junit.Test
public void endodingTest() {
    String text = "07:10Ο Σκούμπι Ντου & ο κολλητός....";
    StringBuilder builder = new StringBuilder();
    for (int i = 0; i < text.length(); i++) {
         char c = text.charAt(i);
         if(Character.isDigit(c)) {
             builder.append("\n");
         }
         builder.append(c);

    }
    System.out.println(builder.toString());
}

The problem is, that everytime you insert a additional \n in the String builder, every char after that line break becomes moved one char backwards. To correct this you have to count all the linebreakes you already inserted and if you insert a new one you have to insert it at position i + numberOfAlreadyInsertedLineBreaks

(builder.insert(i + numberOfAlreadyInsertedLineBreaks, "\n");) complete example below)

the second thing of course (but you already know it) is that you have to improve your pattern, so at the end this is the soultion

@org.junit.Test
public void endodingTest() {
    String text = "07:10Ο Σκούμπι Ντου & ο κολλητός του07:30Πρωϊνή μελέτη10:15Νηστικοί πράκτορες11:15Σαρίτα, είσαι η ζωή μου12:50Οι ειδήσεις του Star13:45Made in Star15:45Μίλα17:45Ειδήσεις17:50Φώτης - Μαρία live19:45Οι ειδήσεις του Star21:00Ο Χαρί Πότερ και ο ημίαιμος πρίγκιψ00:15Σχολή για απατεώνες01:15Supernatural";
    StringBuilder builder = new StringBuilder(text);
    int numberOfAlreadyInsertedLineBreaks = 0;
    for (int i = 0; i < text.length(); i++) {

        if (match(text, i)) {
            builder.insert(i + numberOfAlreadyInsertedLineBreaks, '\n');
            numberOfAlreadyInsertedLineBreaks++;
        }

    }
    System.out.println(builder.toString());
}

private boolean match(String text, int i) {
    return Character.isDigit(text.charAt(i))
            && Character.isDigit(text.charAt(i + 1))
            && text.charAt(i + 2) == ':'
            && Character.isDigit(text.charAt(i + 3))
            && Character.isDigit(text.charAt(i + 4));
}

First, I am afraid you are wrong. I took Greek ABC from Wikipedia and ran the following loop on it:

public static void main(String[] args) {
    System.out.println("before");
    String greek = "ΑαΒβΓγΔδΕεΖζΗηΘθΙιΚκΛλΜμΝνΞξΟοΠπΡρΣσςΤτΥυΦφΧχΨψΩω";
    for (char c : greek.toCharArray()) {
        if (Character.isDigit(c)) {
            System.out.println("digit is found: " + c);
        }
    }
    System.out.println("after");
}

It printed:

before
after

This means that isDigit() works correctly.

Concerning to extracting time from the string. I'd recommend you to use the following regex:

    Pattern p = Pattern.compile("(\\d{2}:\\d{2})");
    Matcher m = p.matcher(str);
    int start = 0;
    while(m.find(start)) {
        String time = m.group(1);
        start = m.end();
                // time variable contain time HH:mm. Just use it as you need
    }

Try Pattern:

scala> java.util.regex.Pattern.compile("(\\d\\d.\\d\\d)").matcher("first12.34second56.78third90.12fourth34.56").replaceAll("$1 ")
res1: java.lang.String = "first12.34 second56.78 third90.12 fourth34.56 "


scala> java.util.regex.Pattern.compile("(\\d\\d.\\d\\d)").matcher("αβγδεζηθικλ12.34αβγδεζηθικλ56.78αβγδεζηθικλ90.12αβγδεζηθικλ34.56").replaceAll("$1 ")
res2: java.lang.String = "αβγδεζηθικλ12.34 αβγδεζηθικλ56.78 αβγδεζηθικλ90.12 αβγδεζηθικλ34.56 "